| Abstract [eng] |
This study analyses relations between stock markets and economic tendencies of BRICS group, which consists of 5 world‘s developing countries. The review of previous scientific researches has revealed that such concept of interaction analysis is not new. Nevertheless, aforementioned investigation does not lose its relevance, because the majority of previous researches mostly have been focused just on developed countries. In addition, recent investigations recognize that developing countries have a significant impact on the development of global economics. Data issues as insufficient access to data and newest information analysis demand remain significant concerns for such research. Therefore, vector autoregression models are used in this study in order to evaluate relations between economic and stock market tendencies for chosen countries. Meanwhile, one-dimensional models for financial time series are applied seeking to evaluate stock market tendencies of BRICS countries. Afterwards, calculations of logarithms and growth rates (returns) are used for modifications of selected stock market and economic tendencies indicators, so the risk of spurious relationships detection could be reduced. It has been noted that components related to variance and shocks of chosen indexes, which reflect stock markets of BRICS countries, significantly affect these time series after the one-dimensional models is set. This means that aforementioned models are more suitable for forecasting the risk rather than the returns of indexes. Moreover, after setting up multivariate (vector autoregression) models, no mutual relations between economic situation indicators and stock indexes have been detected. Furthermore, stock indexes of Brazil, Russia and the Republic of South Africa affects composite leading indicator models while composite leading indicators themselves are significant parameters for stock indexes of India and China. Likewise, some economic indicators such as money supply and inflation leads stock indexes meantime particular economic indicators (exchange rate and inflation) are affected by same stock indexes. Nevertheless, structural analysis shows transient relationships between the stock indexes and the economic indicators of BRICS countries. According to determined forecast errors of predictions on stock indexes, one-dimensional models are more suitable for forecasting of such time series. During the separate investigation of Russia‘s case, the factors to represent stock market‘s indicators are chosen those which contain information of returns for stocks quoted in Moscow Exchange. Thus, more informative and significant vector autoregression models have been obtained. |
